There’s a lot of nonsense being said about the South African Reserve Bank (SARB) review of the prime interest rate. First point: no, this is not about reducing the spread between repo and prime. Sorry to those with prime-linked debt.

The prime rate is a convention. That means it is a reference point that everyone understands. It allows people to compare easily quotes from different lenders. It is not a mechanical pricing mechanism. In fact, banks price around the prime rate, so in reality, what a bank actually charges is different to prime (so prime-linked loans are “prime plus x” or “prime minus x”).

Factors that affect pricing are:

Economic conditions;
Bank risk appetite (which in turn is affected by features of its balance sheet – the balance between maturities, loan-to-value ratios in the rest of the portfolios, etc);
Competitive pressures (yes, these do exist – banks keenly watch what rates people can get elsewhere, including from non-banks);
Statutory capital costs (regulators enforce capital ratios that require banks to hold capital against every loan they make – a different amount depending on the kind of loan);
Shareholder hurdle rates (shareholders expect banks to earn a margin on their money, otherwise they wouldn’t give it to them);
The overall cost of funds (competition, bond market pricing, exchange rates and the like also affects the cost of funds). The higher the cost of funds the more banks must charge to lend; and
The riskiness of the borrower.

Falling in line

The convention since 2001 has been that banks stick to 3.5 percentage points above repo. This isn’t that unusual – other countries have similar conventions, most notably the US, which has prime three points above the Fed target rate.

Before 2001 things were a bit more flexible. Banks determined their own spread above prime. But round about then, I remember an incident when one bank publicly advertised that it would not increase its prime rate following a SARB rate hike (this in the wake of the small banks crisis – those were sensitive times). This went down pretty badly at the central bank; there were stern words with those bank executives and they fell in line. Everyone has stayed in line since.

The SARB would want everyone to be in line to protect the monetary transmission mechanism – the tightness of the SARB’s repo rate decisions and the actual cost of debt in the economy. It would undermine monetary policy if banks ignored repo in their rate pricing.

But, dear reader, this is largely convention. In fact, the repo rate has limited impact on banks’ actual cost of funds, because banks don’t actually borrow (or now deposit) that much at the repo window from the SARB. They have much cheaper sources of funding – such as your lazy deposits on which they pay less than repo.

In fact, in 2022 there was a major overhaul of how repo worked. Until then, the SARB effectively forced banks to borrow at repo once a week. But since then, repo is actually paid on surplus cash the banks deposit with the SARB. They made this change because Covid caused a liquidity spike (cash deposits sitting with banks) so they didn’t need to borrow at repo, which meant the repo rate did not affect their cost of funds at all.

So now the effect of the repo rate is to affect how much banks earn on their excess funds – when it goes up, banks are more likely to put it on deposit; when it goes down, banks are more likely to use it to lend into the economy. This makes for a more effective monetary transmission mechanism, but it is still rather blunt. Repo certainly has an influence on the shorter end of the yield curve (i.e. the funding for banks in the spot market) but it is only one influence among others.

Now, back to the issue of the spread between repo and prime. It is a comforting feature of the convention that the 3.5% is not that far off the actual interest rate spread most banks earn. For the June 2025 results, for instance, these ranged from 3.9%-4.9% among the big four. This is again just convention, though it is somewhat satisfying to common sense that there is a correlation.

Note also that prime is not the only convention banks use. It is common for home loans and other asset finance in retail and small business lending, but larger-scale lending often references market rates like Zaronia (the replacement to Jibar). Those rates are a better benchmark for banks because they reflect what funds actually cost them (Zaronia is the overnight unsecured rate in money markets) and what they can earn in the market.

A benchmark for pricing loans

Why is the SARB interested in reviewing prime? There does remain a lingering concern about the tightness of the monetary transmission mechanism and the signalling effect of the repo.

The SARB wants this to be very clear and direct so that it has a tight rein on the economy, able to accelerate or decelerate through its monetary policy committee decisions. There may be a case that if everything was priced against the repo (for example, “repo plus x%”) then the signalling would be more direct.

The SARB may also want to turn the 3.5 percentage point margin from a convention to a regulation, just to limit risk that some future bank may want to try break with convention. But this seems a pretty minor point.

What I would not expect is any attempt to fiddle with how banks price relative to the benchmark, as that would be wading into price setting and dampening competitive dynamics between banks.

I can’t imagine there’s going to be a different conclusion from the last time a review was undertaken, in 2009. It found, in summary: “The main conclusion of the report is that the size of the spread between the repo rate and prime is immaterial to the setting of lending rates, as prime is primarily used as a reference rate or benchmark for pricing loans. Though it could cause some short-term problems and disruption with existing agreements, any change in the spread or benchmark rate will not change the methodology for establishing actual bank lending rates. A uniform spread helps to create a competitive environment for banks, which enables customers to choose between products and negotiate interest rates based on their credit profile.”

Sixteen years on, that conclusion remains correct: changing the benchmark does not change the economics of lending. It may alter optics and signalling, but it will not reduce borrowing costs in any durable way.

Stuart Theobald is executive chair of Krutham.